Anionic N-heterocyclic carbene ligands from mesoionic imidazolium precursors: Remote backbone arylimino substitution directs carbene coordination

Article abstract of DOI:10.1002/chem.201203488

Versatile precursor to metal complexes: The lithiated N-heterocyclic carbene derivative [Li(imino-imidide)(tmeda)] (imino-imidide=4-(2,6- diisopropylphenylimino)imidazol-2-ylidenide), obtained by deprotonation and lithiation of a mesoionic 4-aryliminoimidazolium, serves as a versatile precursor to metal complexes with the anionic 4-aryliminoimidazolide by salt metathesis. DFT calculations demonstrate that tuneable electronic and steric factors of the substituent at the N_exo imino function of the heterocycle dictate the preferred site of the metallation (C_NHC vs. N_exo). Copyright

Full text of DOI:10.1002/chem.201203488

DOI: 10.1002/chem.201203488
Anionic N-Heterocyclic Carbene Ligands from Mesoionic Imidazolium
Precursors: Remote Backbone Arylimino Substitution Directs Carbene
Coordination
Andreas A. Danopoulos,* Kirill Yu. Monakhov, and Pierre Braunstein*[a]
Dedicated to Professor Willie Motherwell on his retirement
450
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 450 – 455

COMMUNICATION
There is a rapidly growing interest in the chemistry of N-
heterocyclic carbenes (NHC) bonded to early- and middle-
transition metals, main-group elements, lanthanides, and ac-
tinides.[1] This stems from intriguing observations on stabili-
sation of unusual coordination numbers, geometries, oxida-
tion states, and bonding patterns[2,3] and a number of still
unsettled metal–NHC bonding issues.[4,5] Anionic ligands
confer higher stability to their complexes than neutral
donors and this feature attracted us to anionic NHC li-
gands.[6] Known and plausible approaches towards such li-
gands are summarised in Scheme 1. They include functional-
alities to one of the backbone atoms (remote from the CNHC
donor, structures D–F) facilitates interaction with the p-
electron system of the heterocyclic ring and may lead to
electronic tuning of the donor properties of the carbene
function. It may also lead to “Janus-type” ligand coordina-
tion.[9] However, only a limited number of formally anionic
NHC and NHC related ligands are known, either type D,
formally originating from pyrimidine betaine,[10–12] type E,
from diazadiborabenzene[13] or type F from imidazolium
mesoionic frameworks.[14] Anionic NHC ligands of type F
with Y=OÀ or iPrNÀ have been generated in situ but not
fully characterised.[15]
Relevant NHC-proligands (some of them being involved
in tautomeric zwitterion-NHC equilibria)[15b,16] and mesoion-
ic NHC ligands have been recently prepared,[17,18] and as-
pects of their coordination chemistry reported;[9] they over-
all act as neutral ligands. The scarcity of information on the
coordination behavior, scope, and tuning of anionic NHC li-
gands of type F (Y=anionic alkylamido or arylamido sub-
stituents) prompted our research in the area, with the long-
term aim to access stable complexes without using chelating
ligands as in type B. It is conceivable that a remote anionic
amido functionality may incur an additional and/or compet-
ing metal-binding site (for similar behavior in pyrimidine
betaine-derived anionic NHC ligands, see references [10–
12]) leading to ligand polytopicity, ambidenticity or bimetal-
lic coordination.[9,12a,19] This potential extension is not antici-
pated in the boron backbone substituted ligands of type F.[14]
Initially, attempts were made to establish the viability of
well-defined, anionic NHC ligand-transfer reagents (type F,
Y=ArNÀ; Ar=DiPP, 2,6-diisopropylphenyl) and, subse-
quently, to demonstrate their reactivity with metal precur-
sors. N-arylamino functionalisation was chosen to take ad-
vantage of the steric and electronic tuning possibilities in-
herent to the aromatic ring substituent at the Nexo. Our re-
sults disclose the versatility of the ligand design and, with
the help of density functional theory (DFT), provide an un-
derstanding of the scope associated with the choice of the
remote Nexo substituent. Synthetic transformations are out-
lined in Scheme 2 and details are given in the Supporting In-
formation.
Scheme 1. Schematic summary of anionic ligands with NHC donors.
isation of the heterocyclic ring by an anionic group attached
to one nitrogen atom either directly (type A) or through a
linker (type B) or to one carbon backbone atom (type C).
Furthermore, the negative charge could be delocalised over
the backbone atoms of the heterocycle (types D and E).
Currently, the most established approach comprises ligands
of type B, in which hard anionic donors (e.g., ROÀ, R2NÀ
etc.) participate in the formation of multi-dentate chelating
architectures.[6,7] Attachment of two or three NHCs to an
anionic non-coordinating boron centre leads to ligands of
type A (analogous to di- and tris-pyrazolylborates) that can
coordinate to the metal solely through the 2e donor
NHC(s).[8] In contrast, direct attachment of anionic function-
The 4-(2,6-diisopropylphenyl)-substituted imidazolium
chloride 1 was obtained in one step and selectively mono-
deprotonated with one equivalent of KNACHTUNTGRNEUNG(SiMe3)2 in toluene
to give the light yellow zwitterion 2 (see Supporting Infor-
mation for their crystallographic data).[27] Importantly, the
1H NMR spectrum of 2 exhibits two sharp doublets at room
temperature at d=6.00 and 4.92 ppm assignable to the
NCHN and NCHC protons of the heterocyclic ring and
three septets and six doublets due to the isopropyl groups of
[a] Prof. Dr. A. A. Danopoulos, Dr. K. Yu. Monakhov,
Prof. Dr. P. Braunstein
1
the DiPPs. The appearance of the H NMR spectrum pro-
Laboratoire de Chimie de Coordination
Institut de Chimie (UMR 7177 CNRS)
Universitꢁ de Strasbourg
4 rue Blaise Pascal, 67081 Strasbourg Cedex (France)
E-mail: danopoulos@unistra.fr
vides evidence that, in solution, a single zwitterionic tauto-
mer is present, and this is to be contrasted to the tautomeric
equilibrium between the amino–NHC and the zwitterion
tautomers observed when R=iPr.[15b] (Scheme 3). Com-
pound 2 does not exhibit any sensitivity to moisture, in fact
it can be crystallised from wet toluene as the monohydrate.
braunstein@unistra.fr
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201203488.
Chem. Eur. J. 2013, 19, 450 – 455
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
451

P. Braunstein, A. A. Danopoulos et al.
Scheme 2. The synthesis of the anionic NHC proligands and complexes
with arylamino/arylimino remote backbone substituent; Ar and R=
DiPP. Canonical forms with localised p systems are shown for simplicity,
even though the anionic charge is delocalized over the ligand framework.
The metallated–zwitterionic heterocycles in 3 and 4 are anionic ligands,
bound to Li+ and Fe2+, respectively.
Figure 1. The molecular structure of 3; ellipsoids are at 30% probability
level. One molecule of toluene in the asymmetric unit is omitted for
À
À
clarity. Selected bond lengths [ꢂ] and angles [8]. Li1 C1 2.093(3), C1
À
À
À
N1 1.3511(18), C1 N2 1.3840(18), C2 N1 1.3964(18), C3 N2 1.4256(17),
À
À
À
C2 C3 1.384(2), C3 N3 1.3257(18), C28 N3 1.4085(19); C3-N3-C28
116.19(12), N1-C1-N2 102.08(11), N1-C1-Li1 129.47(13), N2-C1-Li1
128.45(13).
Scheme 3. Possible tautomeric equilibrium involving the meso-ionic and
the amino NHC species.[15b] For 2 (Ar, R=DiPP) only the zwitterionic
1
form is detected by H NMR spectroscopy.
Scheme 4. The anionic NHC ligand and its valence isomeric amido car-
bene (Ar, R=DiPP).
Aiming to access an anionic ligand, we reacted 2 with
LiCH2SiMe3 in THF which cleanly afforded an orange mi-
crocrystalline, extremely sensitive solid that after the addi-
tion of tmeda gave crystals of 3·C7H8 from toluene (tmeda=
N, N, N’, N’-tetramethylethylenediamine).
for complexes of type F (Scheme 1) with Y=BEt3, B
ACHTUNGTRENNUNG(C6F5)3,
M=Li(tmeda) or Li
N
ACHTUNGTRENNUNG
1
Characteristically, in the H NMR spectrum of 3, the two
with three-coordinate Li–tmeda alkyls[20] (average 2.122 ꢂ).
Thus, 3 can be viewed as being composed of an anionic
NHC ring functionalised in the backbone position by an ary-
limino group or as a neutral NHC with an anionic amido
substituent at a backbone position. Considering the ob-
served metrical data, 3 could also be described as a one-het-
eroatom-substituted diamino Alder-type carbene[21] linked
doublets mentioned previously for 2 have given way to one
singlet, indicating that further overall deprotonation had oc-
curred on the ring. The presence of Li in solution was sup-
7
ported by Li NMR spectroscopy; however, in the 13C{1H}
NMR spectrum a signal assignable to CNHC was not observa-
ble. The structure of 3 was unequivocally established crystal-
lographically (Figure 1, Scheme 4).[27]
À
through two C N single bonds to an anionic azaallyl struc-
As seen in Figure 1, the Li
N
ture (see below for the discussion comparing metrical data).
heterocycle through the CNHC. Based on the asymmetric unit
contents and the experimental location of relevant hydrogen
atoms of the molecule, the ligand is described as mono-
The deprotonation of 2 could also be achieved by using
KNACTHNUTRGNEUNG
(SiMe3)2 in [D8]THF. The 1H NMR spectrum of the
product was very similar to that of 3; however, in this case
the CNHC signal in the 13C{1H} NMR spectrum was observed
at d=202.3 ppm. We cannot yet conclude whether the K+ is
coordinated to the CNHC (in analogy with the Li in 3) or is
involved in a solvent-separated ion pair with the anionic
NHC ligand or, less likely, coordinated to the exo-nitrogen
atom. Further work in this direction is in progress.
À
À
anionic (cf. Me in MeLi). The CNHC N, the backbone C2 C3
À
and the C3 N3 bonds have partial double-bond character
À
À
(C1 N1 1.3511(18), C1 N2 1.3840(18) ꢂ). The rather short
C28 Nexo bond is indicative of electron delocalisation over
the DiPP ring. The Li CNHC bond length (2.093(3) ꢂ) is
À
À
close to the recently reported values (2.092(2)/2.094(3) ꢂ)
452
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 450 – 455

Anionic NHC Ligands
COMMUNICATION
The consequences of the deprotonation and metallation
of 2 to give 3 (Scheme 4) and the influence of the R sub-
stituent at Nexo on possible tautomeric equilibria (Sche-
me 3)[15b] were explored by DFT calculations at the BP86/
TZ2P level. In particular, it was desirable to establish possi-
ble influence of the Nexo substituent R on the preference for
the formation of C-bound versus Nexo-bound H+ and Li-
linked steric and electronic factors are at work as detailed
below.
Steric effects associated with complexation of LiACTHNUTGRENUNG
(tmeda)+
to Nexo of the anionic NHC ligands were evaluated as a func-
[23]
tion of R by calculation of the %Vbur for the optimised
structures of 1) the amido–imidene anions (i.e., no Li-
AHCTUNGTRENNUNG
(tmeda)+ coordination) and 2) the amido–imidene–Li-
A
ACHTUNGTRENNUNG
involved: 1) the optimisation of the structures of the corre-
sponding C and Nexo valence-bond- and regioisomers and
tautomers; 2) the calculation of their relative (DErel) and
bond dissociation (De) energies for R=Me, iPr, tBu, Ph,
and DiPP; 3) the probing of any steric effects inherent to
AHCTUNGTRENNUNG
%Vbur fall in the range 56–69 and 47–57% and increase in
the order Me<iPr<Ph<DiPP<tBu and Me<Ph<DiPP<
iPr<tBu, respectively. In contrast, the CNHC binding site is
sterically more open in the anionic NHC ligands, with a
the Li
ACHTUNGTRENNUNG
place of LiACHTUNGTRENNUNG
Schemes S1–S4, Supporting In-
formation and discussion on
computational methodology).
As a result, for all R substitu-
ents we observed that H+ pre-
fers to be C-bound by approxi-
mately 18–30 kJmolÀ1 when
R=Me, iPr and approximately
43–53 kJmolÀ1 when R=Ph,
DiPP (Scheme S1 in the Sup-
porting Information). This is in
agreement with our experimen-
tal observations (see also
Scheme 3). In contrast, Li+ pre-
fers to be Nexo-bound by ap-
proximately 50 kJmolÀ1 when
R=Me and iPr and approxi-
mately 45 kJmolÀ1 when R=Ph
and DiPP (Scheme S2 in the
Supporting Information), al-
though in the latter case addi-
tional stabilising interactions
with adjacent aromatic rings
were also detected in the opti-
mised structures (Scheme S3 in
the Supporting Information).
The observed H+ and Li+ site
preference is consistent with
the high affinity of H+ for
[22]
CNHC and carbanions, and of
Li+ for amide (NH2À; Tables S3
and S4 in the Supporting Infor-
mation).
Interestingly, LiACHTUNGTRENNUNG
(tmeda)+ is
calculated to prefer Nexo bind-
ing by approximately 4–
13 kJmolÀ1 when R=Me and
iPr, but CNHC binding by
approximately
when R=Ph, DiPP and tBu
(Scheme S4 in the Support-
ing Information). Here, inter- when going left to right.[24]
4–20 kJmolÀ1
Figure 2. Distribution of the Voronoi deformation density (VDD) atomic charges, Mayer bond orders and rele-
vant highest occupied molecular orbitals (Æ0.03 isosurface value) of the amino–NHC, the anionic amido–imi-
dene ligand, and the Nexo- and CNHC-bound LiACTHNUTRGNE(NUG tmeda) complexes for R=DiPP showing the orbital evolvement
Chem. Eur. J. 2013, 19, 450 – 455
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
453

P. Braunstein, A. A. Danopoulos et al.
minor influence of R (%Vbur =ca. 43–50%, MeꢀiPrꢀtBu<
Ph<DiPP). This is also the case for CNHC after Li
(tmeda)+
AHCTUNGTRENNUNG
coordination (%Vbur =ca. 39–43%, Ph<tBuꢀMeꢀiPr<
DiPP). The comparison between the sterics around Nexo and
CNHC binding sites explains the preference of LiACTHNUTRGNEUNG
(tmeda)+ to
bind to CNHC rather than to Nexo on replacing R=Me or iPr
with tBu because of unfavorable deformations of steric
origin in the latter. In the tBu case, even nonsymmetrical
À
Li–tmeda chelates are observed (see DACTHNUGTRNEUNG(Li N)tmeda in Fig-
ure S12 and calculated deformation energies, DEdef
, in
Table S5 of the Supporting Information).
The differences in the relative energies for R=Ph, DiPP,
and iPr should be ascribed to the contribution of steric and
electronic factors to the deformation energies of the amido–
imidene fragment upon coordination. The fact that Li-
ACHTUNGTRENNUNG
(tmeda)+ complexation results in the reversal of the order
of %Vbur as a function of R points to a substantial readjust-
ment of metrical parameters in order to accommodate the
metal fragment with its energetic consequences. Conforma-
tional changes involving the aromatic ring at the Nexo (for
example, but not exclusively, the torsional angle formed by
the C3-Nexo-C28 plane and the DiPP ring) are indeed ex-
pected to influence the p delocalisation on the ring and the
remaining ligand framework. Extensive p-electronic delocal-
isation over the whole amido–imidene framework is consis-
tent with metrical data for the optimised and experimental
structures (see Figure 2 for frontier MOs and Mayer bond
orders for R=DiPP and Figures S7 and S8 in the Support-
ing Information). This, in combination with the reduced
Figure 3. The molecular structure of 4; ellipsoids are at 30% probability
level; one THF molecule in the asymmetric unit is omitted. Selected
À
À
bond lengths [ꢂ] and angles [8]. C1 Fe1 2.090(2), N4 Fe1 2.1775(19),
À
À
À
À
N5 Fe1 2.209(3), Cl1 Fe1 2.2682(8), C1 N1 1.339(3), C1 N2 1.396(3),
À
À
À
À
À
C2 N1 1.389(3), C3 N2 1.413(3), C2 C3 1.390(3), C3 N3 1.327(3), C28
N3 1.412(3); N1-C1-N2 103.09(19), C1-Fe1-Cl1 115.17(7), N4-Fe1-Cl1
99.59(6), N5-Fe1-Cl1 102.96(8), C1-Fe1-N4 124.91(9), C1-Fe1-N5
125.85(10), N4-Fe1-N5 81.44(8), C3-N3-C28 117.7(2).
In order to compare the structures of the anionic imino–
imidide moieties in 3 and 4 with that of a neutral amino–
NHC moiety, we prepared the AgI complex 5 by the reac-
tion of Ag2O with 1 (Scheme 2).
electrophilicity of LiACHTNUTGRNEUNG
(tmeda)+ compared to Li+ (compare
VDD charges of +0.33 and +1.00, respectively) results in a
weaker Li-Nexo-amido bond that cannot compensate for the
energy demands of the conformational changes implied
above (see calculated bond dissociation energies, De, in
Table S6 of the Supporting Information).
Complex 5 (Figure 4) comprises a linear two-coordinate
Ag centre with a typical Ag CNHC bond length.[26] The dif-
À
The idea of using 3 as a ligand-transfer reagent was also
examined. In view of the current high interest in iron chem-
istry, 3 was reacted with [{FeACTHNUTRGENNGU(m-Cl)ClACHTUNGTRENN(UGN tmeda)}2] to access
novel Fe–NHC complexes of catalytic potential (Scheme 2).
Extremely air- and moisture-sensitive green crystals of
4·THF were obtained and characterised by spectroscopic
and diffraction methods (see Figure 3 and the Supporting
Information).[27]
The tetrahedral FeII is bound through CNHC to the anionic
heterocyclic ligand and additionally to a chloride. Interest-
ingly, the Fe CNHC bond is shorter than in other FeII mono-
À
dentate NHC complexes of Fe (2.122(2)–2.177(3) ꢂ) and
other anionic alkyls and aryls on tetrahedral Fe (2.111(3)–
2.168(3) ꢂ), but slightly longer than in alkyls of three-coor-
dinate Fe (2.056(2)–2.063(3) ꢂ).[2,25] Compared to the metri-
cal data of the ligand framework in 3, there is elongation of
À
the CNHC N bonds in 4, possibly implying reduced p dona-
tion to the CNHC pz orbital and increased interaction of the
pz orbital with Fe dp orbitals. Consistent with the predomi-
Figure 4. The molecular structure of 5; ellipsoids are at 30% probability
level. The N3-H and the methyls of DiPP groups are omitted. Selected
À
À
nantly single-bond character of N1 C2 and N2 C3, the met-
rical data of the azaallyl-type backbone moiety of the ligand
are virtually identical in 3 and 4.
À
À
bond lengths [ꢂ] and angles [8]. C28 N3 1.4085(19), C25 N1 1.347(2),
À
À
À
À
C25 N2 1.368(2), C25 Ag1 2.0735(16), C27 N3 1.375(2), C26 C27
1.355(2); C25-Ag1-Cl1 175.65(4), N1-C25-N2 104.09(13).
454
www.chemeurj.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 450 – 455

Anionic NHC Ligands
COMMUNICATION
[7] S. T. Liddle, I. S. Edworthy, P. L. Arnold, Chem. Soc. Rev. 2007, 36,
1732–1744.
[8] a) R. Frꢇnkel, U. Kernbach, M. Bakola-Christianopoulou, U. Plaia,
M. Suter, W. Ponikwar, H. Nçth, C. Moinet, W. P. Fehlhammer, J.
Organomet. Chem. 2001, 617–618, 530–545; b) J. M. Smith, D.
Subedi, Dalton Trans. 2012, 41, 1423–1429.
ferences between the amino–NHC form in 5 and the imino–
À
imidide form in 3 and 4 manifest themselves in the C Nexo
À
and the C C lengths of the heterocycle, which are shorter
and longer in 3 and 4 respectively, relative to 5, in accord-
ance with an azaallylic character in 3 and 4 as opposed to an
enamine in 5.
[9] D. Mendoza-Espinosa, B. Donnadieu, G. Bertrand, Chem. Asian J.
2011, 6, 1099–1103.
In conclusion, we have demonstrated through a combined
experimental and computational study the versatility, and
potential tunability of the anionic 4-arylimino-substituted-
imidazole-based NHC ligands and showed their transferabil-
ity to transition metals. Although the nitrogen versus carbon
binding affinity may depend on the metal-complex frag-
ments and other subtle effects, it is evident that the remote
backbone substitution offers a versatile way to develop
novel anionic NHC ligands. This is currently being explored
in our laboratory by experimental and computational meth-
ods.
[10] V. Cꢁsar, N. Lugan, G. Lavigne, Chem. Eur. J. 2010, 16, 11432–
11442.
[11] V. Cꢁsar, N. Lugan, G. Lavigne, J. Am. Chem. Soc. 2008, 130,
11286–11287.
[12] a) N. Vujkovic, V. Cꢁsar, N. Lugan, G. Lavigne, Chem. Eur. J. 2011,
17, 13151–13155; b) M. G. Hobbs, C. J. Knapp, P. T. Welsh, J.
Borau-Garcia, T. Ziegler, R. Roesler, Chem. Eur. J. 2010, 16, 14520–
14533.
[13] K. E. Krahulic, H. M. Tuononen, M. Parvez, R. Roesler, J. Am.
Chem. Soc. 2009, 131, 5858–5865.
[14] a) Y. Wang, Y. Xie, M. Y. Abraham, P. Wei, H. F. Schaefer, P. von
Raguꢁ-Schleyer, G. H. Robinson, J. Am. Chem. Soc. 2010, 132,
14370–14372; b) S. Kronig, E. Theuergarten, C. G. Daniliuc, P. G.
Jones, M. Tamm, Angew. Chem. 2012, 124, 3294–3298; Angew.
Chem. Int. Ed. 2012, 51, 3240–3244.
[15] a) L. Benhamou, N. Vujkovic, V. Cꢁsar, H. Gornitzka, N. Lugan, G.
Lavigne, Organometallics 2010, 29, 2616–2630; b) V. Cꢁsar, J.-C.
Tourneux, N. Vujkovic, R. Brousses, N. Lugan, G. Lavigne, Chem.
Commun. 2012, 48, 2349–2351.
[16] C. Fꢇrber, M. Leibold, C. Bruhn, M. Maurer, U. Siemeling, Chem.
Commun. 2012, 48, 227–229.
[17] L. Benhamou, E. Chardon, G. Lavigne, S. Bellemin-Laponnaz, V.
Cꢁsar, Chem. Rev. 2011, 111, 2705–2733.
[18] D. Mendoza-Espinosa, G. Ung, B. Donnadieu, G. Bertrand, Chem.
Commun. 2011, 47, 10614–10616.
[19] D. M. Khramov, A. J. Boydston, C. W. Bielawski, Angew. Chem.
2006, 118, 6332–6335; Angew. Chem. Int. Ed. 2006, 45, 6186–6189.
[20] V. H. Gessner, C. Strohmann, J. Am. Chem. Soc. 2008, 130, 14412–
14413.
[21] J. Vignolle, X. Cattoꢈn, D. Bourissou, Chem. Rev. 2009, 109, 3333–
3384.
Acknowledgements
We thank the CNRS for a “Chercheur Associꢁ” fellowship (to A.A.D.).
A.A.D. also thanks the Rꢁgion Alsace, the Dꢁpartement du Bas-Rhin
and the Communautꢁ Urbaine de Strasbourg for the award of a Guten-
berg Excellence Chair (2010–2011). K.Y.M. is grateful to the DFG and
the Gutenberg funding for postdoctoral fellowships. We thank the CNRS,
the Ministꢃre de LꢄEnseignement Supꢁrieur et de al Recherche (Paris),
the Universitꢁ de Strasbourg and the International Center for Frontier
Reseach in Chemistry, Strasbourg (ucFRC, www.icfrc.fr) for financial
support. The UdS High-Performance Computing Centre and the Labora-
toire de Chimie Quantique (UdS) are thanked for the provision of com-
putational facilities, and the Service de Radiocristallographie (Institut de
Chimie, Strasbourg) for the determination of the crystal structures.
[22] M. Liu, I. Yang, B. Buckley, J. K. Lee, Org. Lett. 2010, 12, 4764–
4767.
[23] a) A. Poater, B. Cosenza, A. Correa, S. Giudice, F. Ragone, V. Scar-
ano, L. Cavallo, Eur. J. Inorg. Chem. 2009, 1759–1766; b) S. A. Di-
Franco, N. A. Maciulis, R. J. Staples, R. J. Batrice, A. L. Odom,
Inorg. Chem. 2012, 51, 1187–1200.
[24] Although in the cases examined here the anionic character of the
ligand is mostly expressed at the Nexo and CNHC sites, the residual
charge at C2 in combination with the nodal properties of some mo-
lecular orbitals hint to an azallylic-substituted-NHC character for
the anionic amido–imidene moiety (see Figure 2 for example,
HOMO and the VDD charges on Nexo, CNHC and C2 which account
for 60% of the total).
[25] L. Xiang, J. Xiao, L. Deng, Organometallics 2011, 30, 2018–2025.
[26] J. C. Garrison, W. J. Youngs, Chem. Rev. 2005, 105, 3978–4008.
[27] CCDC-893387 (1·0.5CH2Cl2), CCDC-893388 (2), CCDC-893389
(3·0.5C7H8), CCDC-893390 (4·THF), and CCDC-893391 (5) contain
the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallograph-
ic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
Keywords: imidazolium · iron · lithium · N-heterocyclic
carbenes · silver
[1] a) P. L. Arnold, I. J. Casely, Chem. Rev. 2009, 109, 3599–3611; b) Y.
Wang, G. H. Robinson, Inorg. Chem. 2011, 50, 12326–12337; c) Y.
Wang, G. H. Robinson, Dalton Trans. 2012, 41, 337–345; d) D.
Martin, M. Soleilhavoup, G. Bertrand, Chem. Sci. 2011, 2, 389–399;
e) M. J. Ingleson, R. A. Layfield, Chem. Commun. 2012, 48, 3579–
3589; f) F. E. Hahn, M. C. Jahnke, Angew. Chem. 2008, 120, 3166–
3216; Angew. Chem. Int. Ed. 2008, 47, 3122–3172.
[2] A. A. Danopoulos, P. Braunstein, M. Wesolek, K. Yu. Monakhov, P.
Rabu, V. Robert, Organometallics 2012, 31, 4102–4105.
[3] J. Monot, A. Solovyev, H. Bonin-Dubarle, ꢅ. Derat, D. P. Curran,
M. Robert, L. Fensterbank, M. Malacria, E. Lacꢆte, Angew. Chem.
2010, 122, 9352–9355; Angew. Chem. Int. Ed. 2010, 49, 9166–9169.
[4] H. Jacobsen, A. Correa, A. Poater, C. Costabile, L. Cavallo, Coord.
Chem. Rev. 2009, 253, 687–703.
[5] D. J. D. Wilson, S. A. Couchman, J. L. Dutton, Inorg. Chem. 2012,
51, 7657–7668.
[6] a) S. Conde-Guadano, M. Hanton, R. P. Tooze, A. A. Danopoulos, P.
Braunstein, Dalton Trans. 2012, 41, 12558–12567; b) S. Conde-Gua-
dano, A. A. Danopoulos, R. Pattacini, M. Hanton, R. P. Tooze, Or-
ganometallics 2012, 31, 1643–1652.
Received: September 29, 2012
Published online: December 23, 2012
Chem. Eur. J. 2013, 19, 450 – 455
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
455